Metadata-Version: 2.4
Name: linkinpy-segmentation
Version: 0.0.3
Summary: LinkinPy CLI package for deep-learning based bioimage segmentation
Author: LinkinPy contributors
License-Expression: MIT
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Image Processing
Requires-Python: >=3.10
Requires-Dist: cellpose<4,>=3.1
Requires-Dist: numpy>=1.23
Requires-Dist: packaging>=23.0
Requires-Dist: tifffile>=2023.8.30
Provides-Extra: test
Requires-Dist: pytest>=8.0; extra == 'test'
Description-Content-Type: text/markdown

# LinkinPy Segmentation

`linkinpy-segmentation` exposes deep-learning segmentation methods as
LinkinPy-compatible command-line tools. The first method uses Cellpose `cyto3`
for general cell segmentation, following the mAIcrobe idea of first-use model
download and cached inference, but without a napari GUI dependency.

## Cellpose Segmentation

```bash
linkinpy-cellpose-segment input.tif output_labels.tif --model-type cyto3 --diameter 30
```

The command stores Cellpose models in:

```text
~/.linkinpy/models/cellpose
```

On each run it checks whether the requested model files are already present. If
they are missing, the command asks Cellpose to download them, then runs
inference and writes the segmentation labels as a TIFF image.

## LinkinPy Metadata

This package includes `linkinpy.yaml`, so `linkinpy-parse` can generate richer
GUI metadata for ImageJ/Fiji and napari.

## Development

```bash
uv sync
uv run pytest
```
